TWI419524B - Controlling method and streaming service system - Google Patents

Description

Control method and streaming service system

The present invention relates to a system and method for streaming services, and more particularly to a multiplex system and a method of adjusting its service performance.

With the advancement of technology, mobile phones have more and more functions. For example, in addition to making or receiving calls, a mobile phone can be used as a multimedia player, camera, editor, or even as a web browser. Mobile phones with multiple applications often use the multiplex processing core to handle these various functions.

Since the system resources of a mobile phone (such as computing space) are usually very limited, in order for many functions to be executed simultaneously, the multiplex processing core must prioritize each job. Instant multimedia streaming services usually have a higher priority order than other jobs. When a multiplexed handheld device transmits a multimedia stream to the network, the instant multimedia streaming service adjusts the transmission bit rate according to the network bandwidth. For example, when the network bandwidth is unstable, the instant multimedia streaming service maintains quality of service (QoS) by reducing the transmission bit rate. If there is sufficient and stable network bandwidth, the instant multimedia streaming service will increase the transmission bit rate to provide better multimedia service quality at the cost of increasing the processing time of this work. As a result, lower priority work may be delayed for too long or even impossible.

Figure 1 shows the timing diagram for a general multiplex system. Hypothetical multiplex system The first work and the second work must be performed, where the first work is immediate work and must be completed within M time periods. The second job is not immediate work and has a long working duration. Generally speaking, the multiplex system performs the first work and the second work by different threads and gives priority to the instant work, so that the immediacy work can be completed within a certain period of time. Figure 1 shows four threads A, B, C, and D with priority order A > B > C > D. Thread A represents an interrupt service routine (ISR) that is executed every M time periods and can be used to control the assignment of threads B and C. In this embodiment, threads A, B, and C are considered as the first job, and thread D having the lowest priority order is regarded as the second job. Since the first work in the first figure uses unrestricted resources, it will affect the execution of the second work. For example, when the time scale is indicated as between 2M and 3M, only a small portion of the processing time is left to perform the second work. A similar situation is even more severe when the first job is source/channel coding and the second job is non-instant work (such as file transfer).

Figure 2 shows a timing diagram of a multiplexed multiplexed streaming service system. The threads A, B, and C in the figure are considered as the first job, and the thread D having the lower priority is regarded as the second job. The first work in this embodiment is the source/channel encoding process. All threads A, B, and C must be used to complete the first job in M time periods. During the time scale indicated as 0 to M, the multiplex system assigns the initial coded bit rate based on the multimedia source content. Since different applications have different initial encoding bit velocities, this does not explain the actual value. The first job is completed in M time periods, and the remaining time is used to execute thread D. The first work is completed in M time periods to indicate that it meets the immediate limit of the first job. Therefore, the coded bit rate during the time scale indicated as M to 2M is increased to improve the quality of the end user. Since the first work is done when the time scale is marked 2M, the coded bit rate during the time scale indication of 2M to 3M will increase. During the period from 3M to 4M, since the first job is not completed in the previous M time period, the encoding bit rate will be greater than the immediate demand. Therefore, the coded bit rate between 3M and 4M in the time scale is reduced. If the network quality is stable, the coded bit rate will be maintained at the time when threads A, B, and C occupy all available time, ie, M time periods. Figure 2 shows that the processing time of the multiplexed stream service is proportional to the stream bit rate. Threads A, B, and C take up all processing time and delay work with lower priority, such as the time scale. Thread D after M is not assigned to processing time.

For multiplexed systems with limited processor computing power (such as mobile phones or PDAs), when the external playback device plays multimedia files through wired or wireless channels (such as Bluetooth, GPRS, Wi-Fi), it has low priority. The quality of the thread's service will delay the update of the human interface, freeze the image decoding frame and other non-important work.

In view of this, the present invention provides a worker with a high priority order. A method of withdrawing processing time to maintain the quality of service of a job with a low priority order.

The present invention provides a streaming service system for performing first and second operations in M time units. The streaming service system includes a processor for comparing the priority order of the first work and the second work, and performing the first work, wherein the priority order of the first work is greater than the priority order of the second work, in the first work The execution of the first job is aborted before the completion, and the second work is performed.

Furthermore, the present invention provides a method of performing first and second operations in M time units. First, the priority order of the first work and the second work is compared. If the priority order of the first job is greater than the priority order of the second job, the first job is performed. The execution of the first job is suspended before the first work described above is completed, and the second job is performed. In some embodiments, the ratio between the first duration and the second duration is a constant. In other embodiments the first duration is proportional to the parameter and the second duration is the N time unit. If the first and second operations can be completed in the M time unit, the parameter is increased, where N is less than M, and vice versa.

The above and other objects, features and advantages of the present invention will become more <RTIgt;

Example:

Figure 3 shows the streaming service system performing the first and second operations in M time units. Streaming service system 30 includes a processor 302. The processor 302 generates the first data by processing the first job and transmits the first data to the network 304 at the transmission rate. The processing time of the first job is proportional to the parameters. When the first job can be performed, the processor 302 pauses the first job for N time units, at which point the processor 302 can perform the second job. The processor 302 will perform the first work after the N time unit. The processor 302 adjusts the parameters based on the network bandwidth and the processing time of the first job. In some embodiments, the first operation is multimedia source encoding and channel encoding, and the parameters are encoding bit rate or encoding quality. For example, the first work may be MPEG encoding, convolutional encoding, or Reed-Solomon encoding processing. The coding quality of the MPEG encoder is proportional to the compression ratio of the inversion. The coded bit rate of the channel encoder represents the number of redundant bits added to the original data. The more redundant bits are added, the more precise the protection of the original data, which in turn increases the size of the encoded material. System 30 encodes the source material and transmits the encoded data to the network. Increasing the coding quality or the channel coding bit rate will make the coded data have stronger protection against channel noise, so the end user receiving the coded data can successfully decode the coded data. Data sizes with larger encoding rates are relatively large and require more processing time. Thus, as long as the first job can be completed in M time units (i.e., the multimedia can be played instantly), processor 302 can increment the parameters. Otherwise, processor 302 will lower the parameters. In some examples, when the network bandwidth is less than the encoding rate The parameters will be lowered so that the encoded data can be smoothly transmitted to the network. In some embodiments, the second job can be a human interface, image decoding, file transfer, or other non-immediate work. The quality of service for these non-immediate jobs is maintained in the streaming service system of these embodiments.

4 is a flow chart 40 showing a method of performing first and second operations in M time units in accordance with another embodiment of the present invention. In step S401, the priority order of the first work and the second work is compared. If the priority order of the first job is greater than the priority order of the second job, the first job performs the first duration in step 402. Execution of the first job is aborted after the first duration has elapsed, and the second job is performed for a second duration in step S403. In some embodiments, the ratio between the first duration and the second duration is a constant. In other embodiments, the first duration is proportional to the parameter and the second duration is an N time unit. If the first and second work can be completed in the M time unit, the parameter will increase, where N is less than the number of M, and vice versa.

Figure 5 shows the parameters, the timing diagrams used to illustrate the first and second operations of flowchart 40, and stream service device 30. The time unit from 0 to M is the parameter used to initiate the first job. Since the first job is completed before the M time unit, the parameters of the first job increase during the M to 2M time unit. The parameters of the first job may be the encoding bit rate, the encoding quality, and the like. Increasing the parameters can improve the playback quality of the receiving end. During the duration of the 2M to 3M time unit, the controller searches for the uncompleted packet of the first working time in the previous M time unit (representing that the current encoding bit rate is too high, and the network bandwidth or system computing power cannot be satisfied. Immediate demand), due to This will lower the encoding bit rate and return it to the previous setting. During the time unit 3M to 4M, the controller looks for a packet completed in the M time unit (representing that the instant performance can be satisfied), and thus fine-tunes the encoding bit rate again to improve quality. After the 4M time unit, the parameters can be maintained if the network quality is stable.

Higher coded bit rates result in better quality. When the network bandwidth or server computing power cannot meet the requirements of the encoding bit rate, the user at the receiving end cannot enjoy the instant multimedia service. Therefore, channel encoders typically require the controller to adjust the output bit rate of the source code based on network status or system traffic to match playback quality and immediate limits. When the network quality is not good or the computing power of the server is insufficient, reducing the encoding bit rate can reduce the need for network bandwidth and encoder processing time. Therefore, it is necessary to maintain the immediate limit of the playback end to achieve the playback quality. When the network quality is stable and the computing power of the server is sufficient, the encoding bit rate is increased to meet the immediate limit while providing better playback quality.

The present invention has been described above with reference to the preferred embodiments thereof, and is not intended to limit the scope of the present invention, and the invention may be modified and modified without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

30‧‧‧Streaming service system

302‧‧‧ processor

304‧‧‧Network

Figure 1 shows the timing diagram for a traditional multiplex system.

Figure 2 shows a timing diagram of a traditional multiplex system with low priority work.

Figure 3 shows the streaming service system used to perform the first and second operations in M time units.

Figure 4 shows a flow chart of the method used to perform the first and second operations in M time units.

Figure 5 shows the timing diagram for the parameters, first and second work.

Claims (14)

A control method for controlling execution of a first job and a second job on a system in a periodic period including an M time unit, comprising: comparing a priority order of the first work and the second work; Performing the first work described above, wherein the priority order of the first work is greater than the priority order of the second work; and the execution of the first work is suspended before the first work is completed, and the second work is performed, wherein Before performing the second work, the first work performs a first duration, the second work performs a second duration, and the ratio between the second duration and the periodic period is a constant. The control method of claim 1, wherein the ratio between the first duration and the second duration is a constant. The control method of claim 1, wherein the first duration is proportional to a parameter, and the method further comprises adding the parameter when the first operation is completed in the M time unit. The control method according to claim 3, further comprising reducing the above parameter when the first work cannot be completed in the M time unit. The control method described in claim 3, further comprising initializing the above parameters, so that the first and second operations are completed in M time units. The control method according to claim 1, wherein the above The first work is implemented by at least two execution sequences, including a first and a second execution sequence, where the first execution sequence is one interrupt service routine executed every M time unit, and the second execution sequence The system is then executed after the first execution sequence described above. The control method of claim 6, wherein the first work is an immediate work, and the second execution sequence is a source code and a channel code. The control method of claim 7, wherein the second work is a human machine interface, image decoding or file transmission. A streaming service system for performing a first and a second work in a periodic period including an M time unit, comprising a processor for comparing a priority order of the first work and the second work, Performing the first work described above, wherein the priority order of the first work is greater than the priority order of the second work, and the execution of the first work is suspended and the second work is performed before the first work is completed, wherein the execution is performed Before the second work, the first work performs a first duration, the second work performs a second duration, and the ratio between the second duration and the periodic period is a constant. The streaming service system of claim 9, wherein the first duration is proportional to a parameter. The streaming service system of claim 9, wherein when the first work and the second work are completed in the M time unit, the processor increases the parameter, and when the first and second work cannot be performed When completed in the M time unit, the above processor lowers the above parameters. The streaming service system of claim 11, wherein the first work is an immediate work. The streaming service system of claim 12, wherein the parameter is a source encoding rate or a channel encoding rate. The streaming service system according to claim 9, wherein the second work is a human machine interface, a video decoding, or a file transmission.